Pulling Stills From Motion

The moving images we see when we watch videos and movies are really just an optical illusion, a trick played by our brain when it’s shown a certain number of images in a short enough time. At about 30 images per second, the motion caused by this stream of still images becomes indistinguishable from the motion we perceive from actual moving objects. That’s why photography and cinematography have always been connected—the tools needed to create moving images are directly descended from still imaging tools, just larger and more powerful. The two technologies have always developed in tandem.

Take the "35mm" format, for example, which became a standard because it’s the same size film as used in traditional movie cameras. In the early days of cinema, you could use a 35mm camera and lens to test the lighting and composition of a scene before capturing it with more expensive film stock.

Through most of the history of imaging, still cameras were still cameras and motion cameras were motion cameras—that has begun to change. Today’s powerful digital video formats have unleashed a new workflow.

This still was pulled from a cooking video, and could easily be used for a cookbook or on social media.

With 4K video, thanks to the large amount of data captured with each frame, it’s now possible to shoot video and extract usable files from single frames. 4K captures around 30 8MB images per second, and the resulting images are good enough to run in print magazines. That means that a photographer can take a 4K camera and shoot video, and then later select individual frames to deliver to a client.

There are some limitations to the 4K-to-still workflow, including a slightly cumbersome series of steps needed to extract individual frames, but the process opens up some amazing possibilities.

Sports photography immediately comes to mind for video-to-still work, thanks to the fast motion of the typical subject. Capture video of a snowboarder in a half-pipe and pick out the best frame from a collection of not 5 or 10 frames per second, but instead from a continuous stream of 30 images each second.

Other types of subjects can benefit from 4K capture, as well. A common dilemma for wedding photographers is the need to capture intimate gestures and looks from a couple during their ceremony, without a constant barrage of shutter clicks. A wedding photographer could easily set a camera on a tripod with a long lens and capture the entire ceremony, cherry-picking the best frames in postproduction.

It’s also not far until 5K video workflows arrive—systems like those from RED already capture video in 6K in order to edit down to 4K in postproduction video workflows. The point at which video moves to 5K is the point at which still photography and videography become indistinguishable.

Some Restrictions May Apply

While a camera capturing 4K video might have a high-megapixel sensor, 4K video doesn’t use all of the data as it’s cropped from the sensor and downsampled to the 4K standard size. Each frame from 4K is an 8.8-megapixel still and is captured at a 17:9 ratio.

The stills grabbed from 4K video will be wider and narrower than a full-frame still. This often will result in the need to crop an image that was captured in 4K when standard still images are also part of the mix, in order to maintain aspect ratios. Cropping the 4K frame grab naturally results in an even lower final pixel count. There’s also reduced dynamic range when shooting video, something that has to be factored in during a shoot.

It’s also not far until 5K video workflows arrive—systems like those from RED already capture video in 6K in order to edit down to 4K in postproduction video workflows. The point at which video moves to 5K is the point at which still photography and videography become indistinguishable.

4K video requires a huge amount of storage space. A minute of 4K video takes about 1 GB of storage space, and high-speed CompactFlash and SD cards are required to keep the camera from filling up the buffer and prematurely ending the video capture. This renders a lot of the CF cards photographers have in their bags useless for 4K video, meaning another outlay for storage.

Photographers will also need additional software to pull images from 4K video—something like Apple’s Final Cut Pro X or Adobe’s Premiere Pro. You can’t just import 4K video into Lightroom and pick a frame (yet), but you can import it into Final Cut Pro X or Premiere Pro, pick a frame and export just that single, 8-megapixel image.

It’s not necessary to have a 4K display to work with 4K video, since the standard automatically scales down to be able to display correctly on HD displays, so there’s no new investment in hardware needed.

Grabbing A Photo

While the process for exporting stills from Premiere Pro and Final Cut Pro X are slightly different, the idea is the same. We’ll assume you’re working with Final Cut Pro X, in this example, but the steps are similar in Premiere Pro.

Final Cut Pro X requires a small bit of configuration to be able to export a frame. From the Preferences menu (press "Command+Comma"), select the Destinations tab and then click Add Destination (Fig. 1). Click on the Save Current Frame icon and select a format. TIFF and Photoshop (PSD) are the highest-resolution options, but PNG and JPEG are also available.

After capturing a 4K video, import it into the application (Fig. 2). You’ll be prompted to add them to a new event or an existing event. If you’re just grabbing stills, it doesn’t matter what event the videos are added to, but if you’re working on a video project, it’s a good idea to give it an appropriate name.

It’s not necessary to turn on any of the special import features, like color balancing or audio analysis, if you’re importing simply to frame grab.

Video clips are added on the left side of the standard Final Cut Pro X interface, and can be left here to grab frames (Fig. 3). It’s not necessary to create a new event or to add videos to an event to select a still.

Simply select the thumbnail of the video and it will appear in the media player; it will have a yellow border to indicate it’s the selected clip.

Scroll the cursor across the thumbnail to select the exact frame or, for more precision, use the right and left arrow keys to step frame by frame through the video. In this example, I was capturing video of a bridge in Portland from a moving boat, trying to capture a frame where the sun created a flare (Fig. 4).

For even more precision, you can add the clip to an event and move the playhead manually.

The final step is to share the still image by clicking on the Share icon on the toolbar. Select the destination for the export and the selected frame will be saved (Fig. 5).

It’s also possible to export an entire range of selected video as a series of frames; this is particularly helpful if you’re trying to illustrate something in sports like snow-boarding in the pipe or skateboarders doing tricks.

The Future Of Stills

It’s possible that, in the near future, many sports, wedding and news photographers will work entirely in a video environment, grabbing stills as necessary from a 4K, 5K or 6K camera. It certainly will become an increasingly viable option for image capture, as it eliminates some of the limitations in capture rate in even the fastest cameras.

In any case, it’s a handy trick for today’s photographer—the ability to silently capture a subject on video and then pull out a usable image can save the day where traditional still imaging is impractical or impossible.

You can reach David Schloss on Twitter or Instagram @davidjschloss 

Leave a Reply